[ef501a]: doc / ocount.1.in  Maximize  Restore  History

Download this file

225 lines (209 with data), 5.8 kB

.\" an page for ocount
.\" Author: Maynard Johnson <maynardj@us.ibm.com>
.TH ocount 1 "@DATE@" "oprofile @VERSION@"
.SH NAME
ocount \- Event counting tool for Linux

.SH SYNOPSIS
.B ocount
[
.I options
]
[ --system-wide | --process-list <pids> | --thread-list <tids> | --cpu-list <cpus> [ command [ args ] ] ]

.SH DESCRIPTION
.BI ocount
is an OProfile tool that can be used to count native hardware events occurring
in either a given application, a set of processes or threads, a subset of active
system processors, or the entire system. The data collected during
a counting session is displayed to stdout by default or, optionally,
to a file.
.P
When counting multiple events, the kernel may not be able to count all events
simultaneously and, thus, may need to multiplex the counting of the events.
If this happens, the "Percent time enabled" column in the
.B ocount
output will be less than 100, but counts are scaled up to a 100% estimated value.
.br

.SH RUN MODES
One (and only one) of the following
.SB run modes
must be specified.  If you run
.BI ocount
using a run mode other than
.BI "command " [args]
, press Ctrl-c to stop
.BI ocount
when finished counting (e.g., when the monitored process ends).
If you background
.BI ocount
(i.e., with '&') while using one these run modes, you
.B must
stop it in a controlled manner so that the data collection process can
be shut down cleanly and final results can be displayed. Use
.BI kill
.BI -SIGINT
.BI <ocount-PID>
for this purpose.
.TP
.BI "command " [args]
The
.I command
is the application for which to count events.
.I args
are the input arguments required by the application.
The
.I command
and its arguments
.B must
be positioned at the
end of the command line, after all ocount options.
.br
.TP
.BI "--process-list / -p " pids
Use this option to count events for one or more already-running applications, specified
via a comma-separated list (
.I pids
). Event counts will be collected for all children of the passed process(es)
as well.
.br

.TP
.BI "--thread-list / -r " tids
Use this option to count events for one or more already-running threads, specified
via a comma-separated list (
.I tids
). Event counts will
.B not
be collected for any children of the passed thread(s).
.br

.TP
.BI "--system-wide / -s"
This option is for counting events for all processes running on your system.  You must
have root authority to run ocount in this mode.
.br

.TP
.BI "--cpu-list / -C " cpus
This option is for counting events on a subset of processors on your system. You must
have root authority to run ocount in this mode. This is a comma-separated list, where each
element in the list may be either a single processor number or a range of processor numbers;
for example: '-C 2,3,4-11,15'.
.br

.SH OTHER OPTIONS
.TP
.BI "--events / -e " event1[,event2[,...]]
This option is for passing a comma-separated list of event specifications
for counting. Each event spec is of the form:
.br
.I "   name[:unitmask[:kernel[:user]]]"
.br
.B Note:
Do
.B not
include a
.I count
value in the event spec, as that parameter is only need when profiling.
.P
.RS
You can specify unit mask values using either a numerical value (hex values
.I must
begin with "0x") or a symbolic name (if the
.I name=<um_name>
field is shown in the
.B ophelp
output). For some named unit masks, the hex value is not unique; thus, OProfile
tools enforce specifying such unit masks value by name.
If no unit mask is specified, the default unit mask value for the event is used.
.P
Event names for certain processor types include a
.I "_GRP<n>"
suffix.  For such cases, the
.I --events
option may be specified with or without the
.I "_GRP<n>"
suffix.
.P
When no event specification is given, the default event for the running
processor type will be used for counting.
Use
.BI ophelp
to list the available events for your processor type.
.RE
.br

.TP
.BI "--separate-thread / -t"
This option can be used in conjunction with either the
.I --process-list
or
.I --thread-list
option to display event counts on a per-thread (per-process) basis.  Without this option, all counts
are aggregated.
.P
.RS
.BI NOTE:
If new threads are started by the process(es) being monitored after counting begins,
the counts for those threads are aggregated with their parent's counts.
.RE

.br
.TP
.BI "--separate-cpu / -c"
This option can be used in conjunction with either the
.I --system-wide
or
.I --cpu-list
option
to display event counts on a per-cpu basis.  Without this option, all counts are aggregated.
.br

.TP
.BI "--time-interval / -i " num_seconds[:num_intervals]
Results collected for each time interval are printed every
.IR "num_seconds"
instead of the default of one dump of cumulative event counts at the end of the run.
If
.I num_intervals
is specified,
.BI ocount
exits after the specified number of intervals occur.
.br

.TP
.BI "--brief-format / -b"
Use this option to print results in the following brief format:
.br
    [optional cpu or thread,]<event_name>,<count>,<percent_time_enabled>
.br
    [        <int>         ,]<  string  >,< u64 >,<     double         >
.P
.RS
If
.I --timer-interval
is specified, a separate line formatted as
.br
    timestamp,<num_seconds_since_epoch>
.br
is printed ahead of each dump of event counts.
.RE

.TP
.BI "--output-file / -f " outfile_name
Results are written to
.I outfile_name
instead of interactively to the terminal.
.br
.TP
.BI "--verbose / -V"
Use this option to increase the verbosity of the output.
.br
.TP
.BI "--version / -v"
Show ocount version.
.br
.TP
.BI "--help / -h"
Display brief usage message.
.br
.TP
.BI "--usage / -u"
Display brief usage message.
.br

.SH EXAMPLE
$ ocount make

.SH VERSION
This man page is current for @PACKAGE@-@VERSION@.

.SH SEE ALSO
operf(1).

Get latest updates about Open Source Projects, Conferences and News.

Sign up for the SourceForge newsletter:





No, thanks